![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to transition.tex CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [CSRG BSD Unix] / 43BSDReno / contrib / isode-beta / doc / nordunet|
Return to transition.tex CVS log | Up to [CSRG BSD Unix] / 43BSDReno / contrib / isode-beta / doc / nordunet |
1.1 root 1: % -*- LaTeX -*- (really SLiTeX)
2:
3: \documentstyle[blackandwhite,landscape,oval,pagenumbers,small]{NRslides}
4:
5: \def\emph#1{\underline{#1}}
6:
7: \font\xx=cmbx10
8: \font\yy=cmbx7
9: \font\sf=cmss10
10:
11: \raggedright
12:
13: \input trademark
14: \let\tradeNAMfont=\relax
15: \let\tradeORGfont=\relax
16:
17: \begin{document}
18:
19: \title {THE OSI CHALLENGE:\\ TRANSITION ASPECTS AND TACTICS}
20: \author {Marshall T.~Rose\\ NYSERNet, Inc.}
21: \date {October 25, 1989}
22: \maketitlepage
23:
24:
25: \begin{bwslide}
26: \part* {OUTLINE}\bf
27:
28: \begin{description}
29: \item[PART I:] MOTIVATION
30:
31: \item[PART II:] BACKGROUND
32:
33: \item[PART III:] PROTOCOL-BASED APPROACHES
34:
35: \item[PART IV:] SERVICE-BASED APPROACHES
36:
37: \item[PART V:] EXAMPLES
38: \end{description}
39: \end{bwslide}
40:
41:
42: \begin{bwslide}
43: \part {MOTIVATION}\bf
44:
45: \begin{nrtc}
46: \item THERE ARE MANY TCP/IP NETWORKS TODAY; THERE WILL BE MORE TOMORROW
47:
48: \item BY THE TIME OSI BECOMES A WORTHWHILE OPERATIONAL ALTERNATIVE,
49: THERE WILL BE MANY MORE TCP/IP NETWORKS THAN THERE ARE TODAY!
50:
51: \item PROBLEM: HOW TO PROTECT INSTALLED BASE?
52:
53: \item PROBLEM: HOW TO TRANSITION GRACEFULLY?
54: \end{nrtc}
55: \end{bwslide}
56:
57:
58: \begin{bwslide}
59: \ctitle {GROWTH OF TCP/IP}
60:
61: \begin{nrtc}
62: \item SALES OF TCP/IP-BASED TECHNOLOGY
63: \begin{nrtc}
64: \item PARTICULARLY IN EUROPE
65: \end{nrtc}
66: CONTINUES TO GROW
67:
68: \item SEVERAL TECHNICAL AND MARKET ASPECTS CONTRIBUTE TO THIS PHENOMENA:
69: \begin{nrtc}
70: \item SUPERIORITY OF TCP/IP IN LOWER-LAYER CONNECTIVITY
71:
72: \item MATURITY OF TCP/IP PRODUCTS\\ (e.g., RANGE OF PLATFORMS)
73: \end{nrtc}
74:
75: \item ALTHOUGH OSI WILL DOMINATE, IT DOESN'T YET
76:
77: \item HENCE, TCP/IP IS BECOMING MORE FIRMLY ENTRENCHED
78: \end{nrtc}
79: \end{bwslide}
80:
81:
82: \begin{bwslide}
83: \ctitle {FEAR AND LOATHING IN THE MARKET}
84:
85: \begin{nrtc}
86: \item F.U.D. IN THE MARKETPLACE:
87: \begin{quote}\em
88: ``All marketing is fear, uncertainty, and doubt.''\\ \raggedleft
89: -- Einar Stefferud, Network Management Associates
90: \end{quote}
91:
92: \item WHAT THE VENDORS SAY:
93: \begin{quote}\em
94: ``$\ldots$ protect your investment while assuring a path to an OSI
95: future.''\\ \raggedleft
96: -- Vendor A
97: \end{quote}
98: AND
99: \begin{quote}\em
100: ``$\ldots$ plans for a smooth, painless guaranteed migration to OSI standards
101: as they are approved.''\\ \raggedleft
102: --Vendor B
103: \end{quote}
104: AND
105: \begin{quote}\em
106: ``Once you've scrapped your existing production networks,
107: come to us for OSI.
108: It will be wonderful!''\\ \raggedleft
109: --Vendor C
110: \end{quote}
111: \end{nrtc}
112: \end{bwslide}
113:
114:
115: \begin{bwslide}
116: \ctitle {THE SAD TRUTH}
117:
118: \begin{quote}\em
119: ``You can't win, and you can't quit, but you \underline{can} reduce the
120: pain.''\\ \raggedleft
121: -- Marshall Rose, NYSERNet, Inc.
122: \end{quote}
123: \end{bwslide}
124:
125:
126: \begin{bwslide}
127: \part {BACKGROUND}\bf
128:
129: \begin{nrtc}
130: \item CONCEPTS
131:
132: \item TERMINOLOGY
133:
134: \item HISTORY
135:
136: \item METRICS FOR COMPARISON
137: \end{nrtc}
138: \end{bwslide}
139:
140:
141: \begin{bwslide}
142: \ctitle {THE FUNDAMENTAL ASSUMPTION}
143:
144: \begin{nrtc}
145: \item TCP/IP IS HERE TODAY, WIDELY INSTALLED, AND USEFUL
146:
147: \item OSI WILL EVENTUALLY REPLACE TCP/IP AS THE OFF-THE-SHELF TECHNOLOGY FOR
148: BUILDING INTEROPERABLE SYSTMS
149:
150: \item BOTH WILL BE SIMULTANEOUSLY WIDESPREAD FOR QUITE SOME TIME
151: \begin{nrtc}
152: \item DURING WHICH OSI WILL GAIN DOMINANCE
153: \end{nrtc}
154: \end{nrtc}
155: \end{bwslide}
156:
157:
158: \begin{bwslide}
159: \part* {CONCEPTS}\bf
160:
161: \begin{nrtc}
162: \item TRANSITION:
163: \begin{nrtc}
164: \item TO MOVE FROM ONE PROTOCOL SUITE TO ANOTHER
165: \end{nrtc}
166:
167: \item COEXISTENCE:
168: \begin{nrtc}
169: \item TO LIVE TOGETHER WITHOUT HOSTILITY OR CONFLICT DESPITE
170: DIFFERENCES
171: \end{nrtc}
172:
173: \item MIGRATION:
174: \begin{nrtc}
175: \item TO MOVE BACK AND FORTH, AS THE SEASONS CHANGE
176: \end{nrtc}
177: \end{nrtc}
178: \end{bwslide}
179:
180:
181: \begin{bwslide}
182: \ctitle {MAPPINGS}
183:
184: \begin{nrtc}
185: \item TRANSITION AND COEXISTENCE CAN BE DESCRIBED BY THE MAPPINGS THEY
186: REQUIRE
187:
188: \item SOME MAPPINGS ARE SIMPLE
189: \begin{nrtc}
190: \item i.e., SYNTACTIC CHANGES
191: \end{nrtc}
192:
193: \item SOME MAPPINGS ARE COMPLEX
194: \begin{nrtc}
195: \item i.e., SEMANTIC CHANGES
196: \end{nrtc}
197:
198: \item THE MORE COMPLEX THE MAPPING, THE GREATER THE LOSS OF INFORMATION OR
199: INTENT
200: \end{nrtc}
201: \end{bwslide}
202:
203:
204: \begin{bwslide}
205: \part* {TERMINOLOGY}\bf
206:
207: \begin{nrtc}
208: \item WE'LL FAVOR OSI TERMINOLOGY, BUT STILL NEED SOME INTERNET (TCP/IP)
209: TERMINOLOGY
210:
211: \item TWO BASIC TERMS
212: \begin{nrtc}
213: \item GATEWAY: GENERIC TO ANY LEVEL, COMPLEX
214:
215: \item BRIDGE: GENERIC TO ANY LEVEL, SIMPLE
216: \end{nrtc}
217: \end{nrtc}
218: \end{bwslide}
219:
220:
221: \begin{bwslide}
222: \ctitle {SERVICE SEMANTICS}
223:
224: \begin{nrtc}
225: \item STORE-AND-FORWARD
226: \begin{nrtc}
227: \item SERVICE SEMANTICS CARRIED MULTI-HOP VIA FORWARDERS
228: \end{nrtc}
229:
230: \item END-TO-END
231: \begin{nrtc}
232: \item SERVICE SEMANTICS CARRIED FROM ORIGINATOR TO RECIPIENT
233:
234: \item MAY BE SUPPORTED BY AN UNDERYLING STORE-AND-FORWARD SERVICE
235: \end{nrtc}
236: \end{nrtc}
237: \end{bwslide}
238:
239:
240: \begin{bwslide}
241: \ctitle {SERVICE SEMANTICS (cont.)}
242:
243: \vskip.5in
244: \diagram[p]{figureT-3}
245: \end{bwslide}
246:
247:
248: \begin{bwslide}
249: \ctitle {PROTOCOL SUITE}
250:
251: \begin{nrtc}
252: \item A COLLECTION OF SERVICES AND PROTOCOLS RELATED:
253: \begin{nrtc}
254: \item ADMINISTRATIVELY, BY AN ORGANIZATION\\ (e.g., ISO/IEC); and,
255:
256: \item PHILOSOPHICALLY, BY A REFERENCE MODEL\\ (e.g., the OSIRM)
257: \end{nrtc}
258:
259: \item FOR OUR PURPOSES, THERE ARE ONLY TWO:
260: \begin{nrtc}
261: \item THE OSI SUITE OF PROTOCOLS
262:
263: \item THE INTERNET SUITE OF PROTOCOLS
264: \end{nrtc}
265: \end{nrtc}
266: \end{bwslide}
267:
268:
269: \begin{bwslide}
270: \ctitle {APPLICATIONS}
271:
272: \begin{nrtc}
273: \item APPLICATION CLASS
274: \begin{nrtc}
275: \item A SET OF APPLICATIONS RELATED TO A PARTICULAR ACTIVITY,
276: e.g., FILE TRANSFER, IRREGARDLESS OF PROTOCOL SUITE
277: \end{nrtc}
278:
279: \item APPLICATION INSTANCE
280: \begin{nrtc}
281: \item A MEMBER OF AN APPLICATION CLASS SPECIFIC TO A PARTICULAR
282: PROTOCOL SUITE, e.g., FTAM
283: \end{nrtc}
284: \end{nrtc}
285: \end{bwslide}
286:
287:
288: \begin{bwslide}
289: \part* {HISTORY}\bf
290:
291: \begin{nrtc}
292: \item A VERY BRIEF INTRODUCTION TO THE TWO PROTOCOL SUITES
293:
294: \item WE'LL ATTEMPT TO TAKE A NON-PARTISAN VIEW (ha!)
295: \end{nrtc}
296: \end{bwslide}
297:
298:
299: \begin{bwslide}
300: \ctitle {INTERNET SUITE}
301:
302: \begin{nrtc}
303: \item SPONSORED BY THE U.S.~DoD
304: \begin{nrtc}
305: \item GREW OUT OF EARLY (D)ARPA RESEARCH INTO SURVIVABLE NETWORKS
306: \end{nrtc}
307: BASIS FROM THE U.S.~DoD INTERNET ARCHITECTURE MODEL
308:
309: \item SPECIFIED IN ``REQUEST FOR COMMENTS'' SERIES (RFCs) AND
310: U.S.~MILITARY STANDARDS (MILSTDs)
311:
312: \item CURRENT GENERATION PRIMARILY BASED ON
313: \begin{nrtc}
314: \item CONNECTION-ORIENTED TRANSPORT SERVICE,
315: PROVIDED BY THE TCP; AND,
316:
317: \item CONNECTIONLESS-MODE NETWORK SERVICE,
318: PROVIDED BY THE IP
319: \end{nrtc}
320:
321: \item MAJOR EMPHASIS ON CONNECTIVITY OF DIVERSE SUB-NETWORKS
322: \begin{nrtc}
323: \item EXCELLENT RESEARCH CONTINUES, TO THIS DAY, ON THESE ISSUES
324: \end{nrtc}
325: \end{nrtc}
326: \end{bwslide}
327:
328:
329: \begin{bwslide}
330: \ctitle {INTERNET SUITE (cont.)}
331:
332: \begin{nrtc}
333: \item SEVERAL PRODUCTION APPLICATIONS
334: \begin{nrtc}
335: \item SIMPLE MAIL TRANSFER PROTOCOL (SMTP)
336:
337: \item FILE TRANSFER PROTOCOL (FTP)
338:
339: \item TELNET (VIRTUAL TERMINAL PROTOCOL)
340:
341: \item DOMAIN NAME SYSTEM (DNS)
342: \end{nrtc}
343: ALL OF WHICH ARE RATHER SIMPLE
344:
345: \item APPLICATIONS CONTAIN THEIR OWN IMPLICIT SESSION AND PRESENTATION
346: MECHANISMS
347:
348: \item NOT SURPRISING, CONSIDERING THAT THESE APPLICATIONS ARE ALL BASED ON
349: 15~YEAR OLD MODELS!
350: \end{nrtc}
351: \end{bwslide}
352:
353:
354: \begin{bwslide}
355: \ctitle {INTERNET PROTOCOLS}
356:
357: \vskip.5in
358: \diagram[p]{figureT-4}
359: \end{bwslide}
360:
361:
362: \begin{bwslide}
363: \ctitle {OSI SUITE}
364:
365: \begin{nrtc}
366: \item SPONSORED BY THE INTERNATIONAL COMMUNITY
367: \begin{nrtc}
368: \item IN PARTICULAR THE ISO
369: \end{nrtc}
370: BASIS FROM THE OSI REFERENCE MODEL (OSIRM)
371:
372: \item SPECIFIED IN ``STANDARDS'' (ISO/IEC) AND RECOMMENDATIONS (CCITT)
373:
374: \item BASED ON
375: \begin{nrtc}
376: \item CONNECTION-ORIENTED TRANSPORT SERVICE,
377: PROVIDED BY ONE OF FIVE DIFFERENT TPs; DEPENDING ON
378:
379: \item THE NETWORK SERVICE AVAILABLE (CONS or CLNS)
380: \end{nrtc}
381:
382: \item DIFFICULT TO IDENTIFY THE ``MAJOR'' EMPHASIS
383: \end{nrtc}
384: \end{bwslide}
385:
386:
387: \begin{bwslide}
388: \ctitle {OSI SUITE (cont.)}
389:
390: \begin{nrtc}
391: \item SEVERAL INTERESTING APPLICATIONS
392: \begin{nrtc}
393: \item MESSAGE HANDLING SYSTEMS (MHS)
394:
395: \item FILE TRANSFER, ACCESS AND MANAGEMENT (FTAM)
396:
397: \item VIRTUAL TERMINAL (VT)
398:
399: \item DIRECTORY SERVICES (DS)
400: \end{nrtc}
401:
402: \item APPLICATIONS EVOLVING QUITE HEAVILY OVER THE LAST FEW YEARS
403:
404: \item MUCH MORE AMBITIOUS THAN THEIR INTERNET COUNTERPARTS
405: \end{nrtc}
406: \end{bwslide}
407:
408:
409: \begin{bwslide}
410: %%%\ctitle {OSI PROTOCOLS}
411:
412: %%%\vskip.25in
413: \diagram[p]{figureT-5}
414: \end{bwslide}
415:
416:
417: \begin{bwslide}
418: \ctitle {A BRIEF COMPARISON}
419:
420: \begin{nrtc}
421: \item NOTE THAT CONCERNS DIFFER
422: \begin{nrtc}
423: \item NETWORK USERS: APPLICATION-LEVEL FUNCTIONALITY
424:
425: \item NETWORK ADMINISTRATORS: NETWORK AND TRANSPORT ISSUES
426: \end{nrtc}
427:
428: \item FOR APPLICATIONS, ONCE IMPLEMENTED, THE OSI SUITE IS SUPERIOR
429:
430: \item FOR NETWORK/TRANSPORT ISSUES, AT PRESENT,
431: THE INTERNET SUITE IS SUPERIOR
432: \end{nrtc}
433: \end{bwslide}
434:
435:
436: \begin{bwslide}
437: \part* {METRICS FOR COMPARISON}\bf
438:
439: \begin{nrtc}
440: \item CAN JUDGE A TRANSITION/COEXISTENCE SCHEME USING DIFFERENT
441: CRITERIA
442:
443: \item THE FOUR WE'LL FOCUS ON ARE ALL SUBJECTIVE;
444: \begin{nrtc}
445: \item TECHNICAL SOLUTIONS DO NOT EXIST IN A VACUUM
446:
447: \item THEY MUST BE EVALUATED IN THE CONTEXT OF A TARGET ENVIRONMENT
448: \end{nrtc}
449: \end{nrtc}
450: \end{bwslide}
451:
452:
453: \begin{bwslide}
454: \ctitle {METRICS FOR COMPARISON (cont.)}
455:
456: \begin{nrtc}
457: \item PERFORMANCE:
458: \begin{nrtc}
459: \item THROUGHPUT, LATENCY
460:
461: \item EFFECT ON OTHER APPLICATIONS
462: \end{nrtc}
463:
464: \item FLEXIBILITY:
465: \begin{nrtc}
466: \item RANGE OF APPLICABILITY
467: \end{nrtc}
468: \end{nrtc}
469: \end{bwslide}
470:
471:
472: \begin{bwslide}
473: \ctitle {METRICS FOR COMPARISON (cont.)}
474:
475: \begin{nrtc}
476: \item TRANSPARENCY:
477: \begin{nrtc}
478: \item USAGE CONTINUITY
479:
480: \item SEAMLESS USER INTERFACE
481: \end{nrtc}
482:
483: \item AMENABILITY:
484: \begin{nrtc}
485: \item MANAGEABILITY
486: \end{nrtc}
487: \end{nrtc}
488: \end{bwslide}
489:
490:
491: \begin{bwslide}
492: \ctitle {SEVERAL CANDIDATES}
493:
494: \begin{nrtc}
495: \item PROTOCOL-BASED APPROACHES
496: \begin{nrtc}
497: \item DUAL STACK
498:
499: \item APPLICATION GATEWAYS
500:
501: \item TRANSPORT GATEWAYS
502: \end{nrtc}
503:
504: \item SERVICE-BASED APPROACHES
505: \begin{nrtc}
506: \item TRANSPORT-SERVICE BRIDGES
507:
508: \item NETWORK TUNNELS
509: \end{nrtc}
510:
511: \item NONE OF THESE TECHNIQUES ARE SPECIFIC TO THE PROBLEM OF
512: \begin{nrtc}
513: \item INTERNET $\mapsto$ OSI
514: \end{nrtc}
515: \end{nrtc}
516: \end{bwslide}
517:
518:
519: \begin{bwslide}
520: \part {PROTOCOL-BASED APPROACHES}\bf
521:
522: \begin{nrtc}
523: \item THE ``STANDARD'' METHODS USED TO INTERCONNECT DIFFERENT
524: PROTOCOL STACKS
525:
526: \item THESE EMPHASIZE THE PROTOCOLS IN EACH STACK
527:
528: \item HENCE THEY REINFORCE THE BOUNDARIES BETWEEN TCP/IP AND OSI
529: \end{nrtc}
530: \end{bwslide}
531:
532:
533: \begin{bwslide}
534: \part* {DUAL STACK}\bf
535:
536: \begin{nrtc}
537: \item PUT BOTH PROTOCOL SUITES IN ALL HOSTS
538:
539: \item WORKS WELL, IF YOU CAN CHANGE EVERYTHING ON THE NETWORK
540: \begin{quote}\em
541: ``Nice work, if you can get it.''\\ \raggedleft
542: -- Groucho Marx, Monkey Business, Paramount Pictures (1931)
543: \end{quote}
544: \end{nrtc}
545: \end{bwslide}
546:
547:
548: \begin{bwslide}
549: \ctitle {DUAL STACK (cont.)}
550:
551: \vskip.5in
552: \diagram[p]{figureT-1}
553: \end{bwslide}
554:
555:
556: \begin{bwslide}
557: \ctitle {TALKING TO UNI-STACK HOSTS}
558:
559: \begin{nrtc}
560: \item QUESTION: HOW TO DECIDE WHICH APPLICATION INSTANCE,
561: \begin{nrtc}
562: \item APPL-$\alpha$ OR APPL-$\gamma$,
563: \end{nrtc}
564: TO USE?
565:
566: \item TWO ANSWERS:
567: \begin{nrtc}
568: \item DEPEND ON THE USER TO KNOW AND INVOKE THE RIGHT PROGRAM
569:
570: \item DEVELOP A GENERIC APPLICATION WHICH SUPPORTS BOTH CLASSES
571: \end{nrtc}
572:
573: \item IN THE LATTER CASE, NEED AN UP-TO-DATE DIRECTORY TO DO THIS RELIABLY
574: \end{nrtc}
575: \end{bwslide}
576:
577:
578: \begin{bwslide}
579: \ctitle {GENERIC APPLICATION INSTANCE}
580:
581: \vskip.5in
582: \diagram[p]{figureT-6}
583: \end{bwslide}
584:
585:
586: \begin{bwslide}
587: \ctitle {AN IMPLEMENTATION OF DUAL-STACK}
588:
589: \begin{nrtc}
590: \item ENVIRONMENT: \unix/~SVR3 (STREAMS)
591:
592: \item ACCESS TO LOWER-LAYER PROTOCOLS VIA TRANSPORT LAYER INTERFACE (TLI)
593:
594: \item NOTE THAT ALTHOUGH TLI PROVIDES A UNIFORM INTERFACE,
595: IT DOES NOT PROVIDE A UNIFORM SERVICE:
596: \begin{nrtc}
597: \item PACKET- vs. STREAM-ORIENTATION
598:
599: \item GRACEFUL RELEASE
600:
601: \item EXPEDITED vs. URGENT DATA
602:
603: \item ADDRESSING
604: \end{nrtc}
605: \end{nrtc}
606: \end{bwslide}
607:
608:
609: \begin{bwslide}
610: \ctitle {GENERIC APPLICATION INSTANCE}
611:
612: \vskip.5in
613: \diagram[p]{figureT-11}
614: \end{bwslide}
615:
616:
617: \begin{bwslide}
618: \ctitle {SCORECARD}
619:
620: \begin{nrtc}
621: \item PERFORMANCE: NO DEGRADATION
622:
623: \item FLEXIBILITY: GOOD
624:
625: \item TRANSPARENCY:
626: \begin{nrtc}
627: \item ASSUMING REMOTE SYSTEM SUPPORTS AT LEAST ONE OF THE PROTOCOL
628: STACKS, THEN HIGH TRANSPARENCY BY USING COMMON SERVICE
629: INTERFACE
630: \end{nrtc}
631:
632: \item AMENABILITY:
633: \begin{nrtc}
634: \item BOTH END- AND INTERMEDIATE-SYSTEMS MUST RUN BOTH PROTOCOLS
635:
636: \item INTRODUCES ADMINISTRATIVE PROBLEMS AS THERE ARE NOW TWO
637: LOGICAL NETWORKS
638: \begin{nrtc}
639: \item MANAGEMENT OF BOTH \underline{PLUS} CONTENTION BETWEEN THEM
640: \end{nrtc}
641: \end{nrtc}
642: \end{nrtc}
643: \end{bwslide}
644:
645:
646: \begin{bwslide}
647: \part* {APPLICATION GATEWAYS}\bf
648:
649: \begin{nrtc}
650: \item A WELL-KNOWN, BUT LITTLE-UNDERSTOOD TECHNOLOGY
651: \begin{nrtc}
652: \item USED IN MESSAGE HANDLING QUITE A BIT
653: \end{nrtc}
654:
655: \item MOST ARE QUITE TERRIBLE
656: \begin{quote}\em
657: ``Sometimes when you try to turn an apple into an orange you get back a
658: lemon.''\\ \raggedleft
659: -- Michael Padlipsky, The Elements of Networking Style (1985)
660: \end{quote}
661: \end{nrtc}
662: \end{bwslide}
663:
664:
665: \begin{bwslide}
666: \ctitle {APPLICATION GATEWAYS (cont.)}
667:
668: \vskip.5in
669: \diagram[p]{figureT-2}
670: \end{bwslide}
671:
672:
673: \begin{bwslide}
674: \ctitle {IMPERFECT MAPPINGS}
675:
676: \begin{nrtc}
677: \item BECAUSE THEY ARE AT THE HIGHEST LAYER IN THE STACK,
678: APPLICATION GATEWAYS TEND TO PERFORM SEMANTIC MAPPINGS
679:
680: \item THESE ARE ACCOMPANIED BY A LOSS OF INFORMATION
681:
682: \item SOMETIMES THE LOSS IS ONLY ANNOYING
683: \begin{nrtc}
684: \item e.g., ``FUNNY LOOKING'' MAIL ADDRESSES
685: \end{nrtc}
686:
687: \item SOMETIMES THE LOSS IS CATASTROPHIC
688: \begin{nrtc}
689: \item e.g., ROUTING LOOPS
690: \end{nrtc}
691: \end{nrtc}
692: \end{bwslide}
693:
694:
695: \begin{bwslide}
696: \ctitle {AN IMPLEMENATION OF APPLICATION-GATEWAY}
697:
698: \begin{nrtc}
699: \item TWO KINDS OF IMPLEMENATIONS
700:
701: \item STAGING (TRUE STORE-AND-FORWARD):
702: \begin{nrtc}
703: \item TOP-LEVEL PROTOCOL TRANSACTIONS ARE GROUPED AT THE GATEWAY
704:
705: \item REQUIRES LOCAL STORAGE, BUT MAY PERMIT BETTER MAPPINGS
706: \end{nrtc}
707:
708: \item IN-SITU (VIRTUAL END-TO-END):
709: \begin{nrtc}
710: \item NO PROTOCOL TRANSACTIONS ARE GROUPED
711:
712: \item MAPPINGS ARE ``ON THE FLY''\\ (AND PERHAPS LESS PRECISE)
713:
714: \item END-TO-END RESPONSE IS FASTER
715: \end{nrtc}
716: \end{nrtc}
717: \end{bwslide}
718:
719:
720: \begin{bwslide}
721: \ctitle {INVOKING THE GATEWAY}
722:
723: \vskip1.5in
724: \begin{verbatim}
725: % ftp file-gateway
726: Name (file-gateway:asterix): obelix@osi-host
727: Password:
728: \end{verbatim}
729: \end{bwslide}
730:
731:
732: \begin{bwslide}
733: \ctitle {A STAGING IMPLEMENTATION}
734:
735: \vskip.5in
736: \diagram[p]{figureT-12}
737: \end{bwslide}
738:
739:
740: \begin{bwslide}
741: \ctitle {AN IN-SITU IMPLEMENTATION}
742:
743: \vskip.5in
744: \diagram[p]{figureT-13}
745: \end{bwslide}
746:
747:
748: \begin{bwslide}
749: \ctitle {SCORECARD}
750:
751: \begin{nrtc}
752: \item PERFORMANCE: USUALLY POOR, BUT ACCEPTABLE FOR STORE-AND-FORWARD
753: APPLICATIONS
754: \begin{nrtc}
755: \item TYPICALLY ALSO INTRODUCES ADDITIONAL NETWORK TRAFFIC
756: \end{nrtc}
757:
758: \item FLEXIBILITY: NONE; EACH A-GWY IS A SPECIAL-PURPOSE SOFTWARE BOX
759:
760: \item TRANSPARENCY:
761: \begin{nrtc}
762: \item TO SERVICE: OFTEN LOSES SIGNIFICANT FUNCTIONALITY
763:
764: \item TO USERS: POSSIBLE, BUT NOT LIKELY (e.g., IN AN FTAM/FTP A-GWY,
765: USERS EMBED HOSTNAMES IN FILENAMES)
766: \end{nrtc}
767:
768: \item AMENABILITY:
769: \begin{nrtc}
770: \item REQUIRES NO END-SYSTEM MODIFICATION
771:
772: \item MAY INTRODUCE ADMINISTRATIVE PROBLEMS
773: \end{nrtc}
774: \end{nrtc}
775: \end{bwslide}
776:
777:
778: \begin{bwslide}
779: \part* {TRANSPORT GATEWAYS}\bf
780:
781: \begin{nrtc}
782: \item IDEA: GATEWAY AT THE TRANSPORT LAYER SO AS TO AVOID NEEDING
783: MULTIPLE APPLICATION GATEWAYS
784: \begin{quote}\em
785: ``We could do it, but it would be wrong.''\\ \raggedleft
786: -- Richard Nixon, The Watergate Tapes (1974)
787: \end{quote}
788:
789: \item ALTHOUGH THE OSI (TP4) AND INTERNET (TCP) TRANSPORT PROTOCOLS DIFFER,
790: THE SERVICE IS QUITE SIMILAR
791:
792: \item HENCE, IT IS TECHNICALLY FEASIBLE TO PERFORM THE MAPPINGS
793: \begin{nrtc}
794: \item (ALTHOUGH IT'S A LOT OF HARD WORK)
795: \end{nrtc}
796: \end{nrtc}
797: \end{bwslide}
798:
799:
800: \begin{bwslide}
801: \ctitle {TRANSPORT GATEWAYS (cont.)}
802:
803: \vskip.5in
804: \diagram[p]{figureT-14}
805: \end{bwslide}
806:
807:
808: \begin{bwslide}
809: \ctitle {THE OBVIOUS QUESTION}
810:
811: \begin{nrtc}
812: \item WHAT APPLICATION DO YOU RUN WHEN USING THIS?
813: \begin{nrtc}
814: \item CAN'T RUN INTERNET APPLICATIONS IN THE OSI NETWORK,
815: SINCE THE TRANSPORT GATEWAY YIELDS OSI TRANSPORT SEMANTICS
816:
817: \item CAN'T RUN OSI APPLICATIONS IN THE INTERNET NETWORK,
818: SINCE THE TRANSPORT GATEWAY YIELDS INTERNET TRANSPORT SEMANTICS
819: \end{nrtc}
820:
821: \item THIS APPROACH FAILS BECAUSE IT PRESENTS DIFFERENT SERVICE SEMANTICS
822: IN EACH NETWORK
823: \end{nrtc}
824: \end{bwslide}
825:
826:
827: \begin{bwslide}
828: \part {SERVICE-BASED APPROACHES}\bf
829:
830: \begin{nrtc}
831: \item BY THE TIME OSI-BASED NETWORKS ARE TRULY WIDESPREAD,
832: TCP/IP-BASED NETWORKS WILL ALREADY OFFER A MIX OF SERVICES:
833: \begin{nrtc}
834: \item SUCH AS FTAM AND MHS, IN ADDITION TO FTP AND SMTP
835: \end{nrtc}
836:
837: \item IN OTHER WORDS, PERHAPS THE TRANSITION TO OSI BEGINS WITH NEW
838: APPLICATIONS ON HOSTS AND NO CHANGES TO THE NETWORK
839: \end{nrtc}
840: \end{bwslide}
841:
842:
843: \begin{bwslide}
844: \ctitle {WOULD THIS REALLY HAPPEN?}
845:
846: \begin{nrtc}
847: \item RECALL THAT USERS ARE INTERESTED IN \underline{SERVICES} NOT
848: \underline{PROTOCOLS}
849:
850: \item THE OSI APPLICATIONS ARE MUCH RICHER THAN THEIR INTERNET COUNTERPARTS
851:
852: \item IN CONTRAST, AT THE LOWER-LAYERS THE INTERNET SUITE ``WORKS BETTER''
853: \begin{nrtc}
854: \item AS SUCH, IT IS UNLIKELY TO BE REPLACED BY THE OSI LOWER-LAYERS
855: FOR QUITE SOME TIME
856: \end{nrtc}
857: \end{nrtc}
858: \end{bwslide}
859:
860:
861: \begin{bwslide}
862: \ctitle {OBSERVATION}
863:
864: \begin{nrtc}
865: \item GIVEN THE ABOVE ASSUMPTION, IT SHOULD BE NOTED THAT:
866: \begin{nrtc}
867: \item WE HAVE TWO COMMUNITIES USING THE SAME APPLICATIONS
868: (OSI), AND
869:
870: \item ONLY THE UNDERLYING ``TS-STACK'' WILL DIFFER BETWEEN THE TWO:
871: \begin{nrtc}
872: \item IN THE OSI COMMUNITY: TP4/CLNP/$\ldots$
873:
874: \item IN THE TCP COMMUNITY: RFC1006/TCP/IP/$\ldots$
875: \end{nrtc}
876: \end{nrtc}
877:
878: \item THIS LEADS US TO POSTULATE AN INTERESTING COEXISTENCE
879: STRATEGY:
880: \begin{nrtc}
881: \item LET'S RUN OSI APPLICATIONS, END-TO-END, BETWEEN THE TWO
882: \end{nrtc}
883:
884: \item IN A SENSE, THIS IS A HYBRID OF THE TWO PREVIOUS APPROACHES,
885: INTENDED TO MINIMIZE THE DISADVANTAGES OF EACH
886: \begin{nrtc}
887: \item SAME APPLICATION PROTOCOL,\\
888: BUT DIFFERENT UNDERYLING LAYERS
889: \end{nrtc}
890: \end{nrtc}
891: \end{bwslide}
892:
893:
894: \begin{bwslide}
895: \part* {TRANSPORT-SERVICE BRIDGES}\bf
896:
897: \begin{nrtc}
898: \item INTRODUCE A TRANSPORT ENTITY CALLED THE ``TS-BRIDGE''
899: \begin{quote}\em
900: ``Users are interested in services, not protocols.''\\ \raggedleft
901: -- Marshall Rose, NYSERNet, Inc.
902: \end{quote}
903:
904: \item ALTHOUGH MANY DIFFERENT TS-STACKS EXIST,
905: THEY ALL PROVIDE THE SAME TRANSPORT SERVICE
906:
907: \item SO, IT IS STRAIGHT-FORWARD TO BUILD A BOX THAT:
908: \begin{nrtc}
909: \item KNOWS NOTHING ABOUT TRANSPORT PROTOCOLS, BUT
910:
911: \item KNOWS HOW TO USE THE RELATIVELY SIMPLE OSI TRANSPORT SERVICE
912: \end{nrtc}
913: \end{nrtc}
914: \end{bwslide}
915:
916:
917: \begin{bwslide}
918: \ctitle {TRANSPORT-SERVICE BRIDGES (cont.)}
919:
920: \begin{nrtc}
921: \item THE TS-BRIDGE ``COPIES'' SERVICE PRIMITIVES FROM ONE TS-STACK TO THE
922: OTHER, e.g.:
923: \begin{nrtc}
924: \item UPON RECEIVING A T-CONNECT.INDICATION PRIMITIVE FROM ONE
925: TS-STACK,
926:
927: \item IT ISSUES A T-CONNECT.REQUEST PRIMITIVE TO THE OTHER TS-STACK
928: \end{nrtc}
929: \end{nrtc}
930: \end{bwslide}
931:
932:
933: \begin{bwslide}
934: \ctitle {TRANSPORT-SERVICE BRIDGES (cont.)}
935:
936: \vskip.5in
937: \diagram[p]{figureT-9}
938: \end{bwslide}
939:
940:
941: \begin{bwslide}
942: \ctitle {CONS vs. CLNS CONNECTIVITY}
943:
944: \vskip.5in
945: \diagram[p]{figureT-19}
946: \end{bwslide}
947:
948:
949: \begin{bwslide}
950: \ctitle {USE OF THE TS-BRIDGE (cont.)}
951:
952: \begin{nrtc}
953: \item ENCODE THE NETWORK ADDRESS AND TRANSPORT SELECTOR AS AN OCTET STRING,
954: CALL THIS THE NEW TRANSPORT SELECTOR
955:
956: \item USE THE NETWORK ADDRESS OF THE TS-BRIDGE FOR THE REMAINING STEPS
957:
958: \item WHEN TS-BRIDGE RECEIVES CONNECTION,
959: IT SIMPLY DECODES TRANSPORT SELECTOR TO FIND ADDRESS OF
960: DESTINATION END-SYSTEM
961: \end{nrtc}
962: \end{bwslide}
963:
964:
965: \begin{bwslide}
966: \ctitle {TS-BRIDGE ADDRESSING}
967:
968: \vskip.5in
969: \diagram[p]{figureT-20}
970: \end{bwslide}
971:
972:
973: \begin{bwslide}
974: \ctitle {THE TS-BRIDGE AND THE OSI MODEL}
975:
976: \begin{nrtc}
977: \item THE TS-BRIDGE IS A LEVEL-FOUR ROUTER
978:
979: \item POTENTIAL PROBLEMS:
980: \begin{nrtc}
981: \item THE TS-BRIDGE MAINTAINS STATE AS TO THE EXISTING CONNECTIONS
982:
983: \item TWO CHECKSUMS, AND NEITHER REALLY END-TO-END
984:
985: \item \underline{MAY} THWART SOPHISTICATED BACK-PRESSURE TECHNIQUES
986: \end{nrtc}
987: \end{nrtc}
988: \end{bwslide}
989:
990:
991: \begin{bwslide}
992: \ctitle {AN IMPLEMENTATION OF THE TS-BRIDGE}
993:
994: \begin{nrtc}
995: \item FIRST DEMONSTRATION IN FEBRUARY, 1988
996: \begin{nrtc}
997: \item TP4/CLNP to RFC1006/TCP
998: \end{nrtc}
999:
1000: \item ANOTHER IMPLEMENTATION IN EUROPE IS HANDLING
1001: \begin{nrtc}
1002: \item TP0/X.25 to RFC1006/TCP
1003: \end{nrtc}
1004: \end{nrtc}
1005: \end{bwslide}
1006:
1007:
1008: \begin{bwslide}
1009: \ctitle {SCORECARD}
1010:
1011: \begin{nrtc}
1012: \item PERFORMANCE: FAIR; WHEN TS-BRIDGE IS MADE INTO A KERNEL-RESIDENT
1013: STREAMS MODULE IT SHOULD IMPROVE DRAMATICALLY
1014:
1015: \item FLEXIBILITY: HIGH; INDEPENDENT OF ANY APPLICATION
1016:
1017: \item TRANSPARENCY: TOTAL
1018:
1019: \item AMENABILITY:
1020: \begin{nrtc}
1021: \item TCP END-SYSTEMS MUST RUN ``NEW'' PROTOCOLS
1022: \begin{nrtc}
1023: \item BUT, NO MODIFICATIONS REQUIRED TO END-SYSTEM KERNELS
1024: \end{nrtc}
1025:
1026: \item MAY INTRODUCE ADMINISTRATIVE PROBLEMS
1027: \end{nrtc}
1028: \end{nrtc}
1029: \end{bwslide}
1030:
1031:
1032: \begin{bwslide}
1033: \part* {NETWORK TUNNELS}\bf
1034:
1035: \begin{nrtc}
1036: \item IDEA: ENCAPSULATE CLNP INSIDE OF IP, TREATING IP AS SIMPLY A DATA LINK
1037: PROTOCOL
1038: \begin{quote}\em
1039: ``Encapsulation complies with the layering concept, but violates the notion
1040: of absolute levels.''\\ \raggedleft
1041: -- Danny Cohen and Jon Postel, ``The ISO Reference Model and Other Protocol
1042: Architectures'' (1983)
1043: \end{quote}
1044:
1045: \item NS-TUNNEL PERFORMS AS A ROUTER, REMOVING ONE DATA LINK HEADER AND
1046: ADDING ANOTHER
1047:
1048: \item METHOD SPECIFIED IN [RFC1070]
1049:
1050: \item ADDRESS MAPPINGS SPECIFIED IN [RFC1069]
1051: \end{nrtc}
1052: \end{bwslide}
1053:
1054:
1055: \begin{bwslide}
1056: \ctitle {TUNNELING}
1057:
1058: \vskip.5in
1059: \diagram[p]{figureT-18}
1060: \end{bwslide}
1061:
1062:
1063: \begin{bwslide}
1064: \ctitle {NETWORK TUNNELS}
1065:
1066: \vskip.5in
1067: \diagram[p]{figureT-10}
1068: \end{bwslide}
1069:
1070:
1071: \begin{bwslide}
1072: \ctitle {INTERESTING FEATURES}
1073:
1074: \begin{nrtc}
1075: \item NO STATE MAINTAINED BY NS-TUNNEL
1076:
1077: \item A TRUE END-TO-END CHECKSUM
1078: \end{nrtc}
1079: \end{bwslide}
1080:
1081:
1082: \begin{bwslide}
1083: \ctitle {POTENTIAL PROBLEMS}
1084:
1085: \begin{nrtc}
1086: \item REQUIRES COMMON HIGHER-LEVEL PROTOCOLS (TRANSPORT AND ABOVE) ON BOTH
1087: END-SYSTEMS, BUT DOES NOT REQUIRE ALL INTERVENING ROUTERS TO USE THE
1088: SAME NETWORK PROTOCOL
1089:
1090: \item THE TCP END-SYSTEM IMPLEMENTATION CHOICES ARE SIMILAR TO NETBIOS OVER
1091: TCP [RFC1001/1002]
1092: \end{nrtc}
1093: \end{bwslide}
1094:
1095:
1096: \begin{bwslide}
1097: \ctitle {AN IMPLEMENATION OF AN NS-TUNNEL}
1098:
1099: \begin{nrtc}
1100: \item HAVEN'T SEE ANY YET
1101: \begin{nrtc}
1102: \item BUT WILL BE IN 4.4BSD UNIX
1103: \end{nrtc}
1104:
1105: \item NEED A LOT OF CLNP-BASED NETWORKS BEFORE THIS IS OF USE
1106:
1107: \item SO THIS WILL HAPPEN AT THE END OF THE TRANSITION PERIOD
1108: \end{nrtc}
1109: \end{bwslide}
1110:
1111:
1112: \begin{bwslide}
1113: \ctitle {SCORECARD}
1114:
1115: \begin{nrtc}
1116: \item PERFORMANCE: NO WORSE THAN TYPICAL CLNP-ROUTER (AND PROBABLY A LOT
1117: BETTER TOO!)
1118:
1119: \item FLEXIBILITY: HIGH (INDEPENDENT OF ANY APPLICATION)
1120:
1121: \item TRANSPARENCY: TOTAL
1122:
1123: \item AMENABILITY: TCP END-SYSTEMS MUST RUN BOTH TRANSPORT PROTOCOLS
1124: \end{nrtc}
1125: \end{bwslide}
1126:
1127:
1128: \begin{bwslide}
1129: \part {EXAMPLES}\bf
1130:
1131: \begin{nrtc}
1132: \item U.S.~DoD OSI IMPLEMENTATION PLAN
1133:
1134: \item GENERIC EXAMPLE
1135:
1136: \item CONCLUSIONS
1137: \end{nrtc}
1138: \end{bwslide}
1139:
1140:
1141: \begin{bwslide}
1142: \part* {U.S.~DoD OSI\\ IMPLEMENTATION PLAN}\bf
1143:
1144: \begin{nrtc}
1145: \item IMPLEMENT CAPABILITY TO USE OSI IN DoD INTERNETWORK ENVIRONMENT
1146: \begin{nrtc}
1147: \item OSI-POSIX PROJECT
1148: \end{nrtc}
1149:
1150: \item PROVIDE THE CAPABILITY FOR U.S.~DoD AND OSI PROTOCOLS TO INTEROPERATE
1151: \begin{nrtc}
1152: \item FTAM-FTP GATEWAY
1153:
1154: \item MHS-SMTP GATEWAY
1155: \end{nrtc}
1156: \end{nrtc}
1157: \end{bwslide}
1158:
1159:
1160: \begin{bwslide}
1161: \ctitle {OSI-POSIX PROJECT}
1162:
1163: \begin{nrtc}
1164: \item GOAL: ACCELLERATE THE UBIQUITY OF OSI
1165:
1166: \item APPROACH: OPENLY AVAILABLE, COMPLETE OSI IMPLEMENTATION FOR NEXT MAJOR
1167: RELEASE OF BERKELEY \unix/
1168:
1169: \item FOR MORE DETAILS:
1170: \begin{quote}
1171: OSI PROTOCOLS WITHIN AN OPENLY AVAILABLE, POSIX-CONFORMANT, BERKELEY UNIX
1172: ENVIRONMENT
1173: \end{quote}
1174: APPEARING IN ConneXions, OCTOBER, 1988
1175: \end{nrtc}
1176: \end{bwslide}
1177:
1178:
1179: \begin{bwslide}
1180: \diagram[p]{figureT-15}
1181: \end{bwslide}
1182:
1183:
1184: \begin{bwslide}
1185: \diagram[p]{figureT-16}
1186: \end{bwslide}
1187:
1188:
1189: \begin{bwslide}
1190: \part* {GENERIC EXAMPLE}\bf
1191:
1192: \begin{nrtc}
1193: \item TWO PRONGS:
1194: \begin{nrtc}
1195: \item FAVOR USE OF OSI APPLICATIONS OVER TCP ON LAN MESH
1196:
1197: \item LOCATE APPLICATION GATEWAYS AND A TS-BRIDGE ON ALL NODES
1198: WITH WAN ATTACHMENETS
1199: \end{nrtc}
1200:
1201: \item AWAIT OSI LOWER-LAYERS TO BECOME COMPETITIVE
1202: \end{nrtc}
1203: \end{bwslide}
1204:
1205:
1206: \begin{bwslide}
1207: \ctitle {GENERIC EXAMPLE (cont.)}
1208:
1209: \begin{nrtc}
1210: \item EACH ATTACHMENT LOCUS SHOULD SUPPORT COEXISTENCE SERVICES
1211:
1212: \item IF RESOURCES PERMIT, SELECT ONE OTHER SYSTEM TO SUPPORT THESE
1213: SERVICES FOR USE BY LOCAL UNI-STACK HOSTS
1214:
1215: \item THIS ``COVERS ALL BASES'' BY HANDLING ALL POSSIBLE OSI COMBINATIONS
1216: WITH A BIT OF EXTRA REDUNDANCY
1217:
1218: \item MIGHT REQUIRE A BIT OF SOPHISTICATED USE FROM THE DIRECTORY
1219: \end{nrtc}
1220: \end{bwslide}
1221:
1222:
1223: \begin{bwslide}
1224: \ctitle {A LAN OF MANY COLORS}
1225:
1226: \vskip.5in
1227: \diagram[p]{figureT-17}
1228: \end{bwslide}
1229:
1230:
1231: \begin{bwslide}
1232: \part* {CONCLUSIONS}\bf
1233:
1234: \begin{quote}\em
1235: ``Optimality differs according to context.''\\ \raggedleft
1236: -- Michael Padlipsky, The Elements of Networking Style (1985)
1237: \end{quote}
1238: \end{bwslide}
1239:
1240:
1241: \begin{bwslide}
1242: \ctitle {CONCLUSIONS (cont.)}
1243:
1244: \begin{nrtc}
1245: \item TCP/IP-BASED NETWORKS WILL OFFER OSI APPLICATIONS
1246:
1247: \item COEXISTENCE IN THE SHORT TERM:
1248: \begin{nrtc}
1249: \item TS-BRIDGE MINIMIZES SOFTWARE INVESTMENT
1250: \end{nrtc}
1251:
1252: \item COEXISTENCE IN THE LONG TERM:
1253: \begin{nrtc}
1254: \item NS-TUNNEL MAXIMIZES PERFORMANCE AND ROBUSTNESS
1255: \end{nrtc}
1256:
1257: \item IF/WHEN THERE ARE NO MORE TCP/IP-BASED NETWORKS, THEN THE
1258: COEXISTENCE PERIOD IS OVER, AND TRANSITION IS A NON-ISSUE!
1259: \end{nrtc}
1260: \end{bwslide}
1261:
1262: \end{document}
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.